Summary Autoimmune disease is clearly linked to aging, but the mechanism by which self-tolerance breaks down with age is not clear. Inducing T cells to become self-tolerant is a function of the thymus, and is thought to involve presentation (directly or through cross-priming) of a broad spectrum of peripheral self-antigens (tissue-restricted antigens, TRA) expressed by medullary thymic epithelial cells (mTEC). We have recently shown that expression of TRA by mTEC decreases with age. Progressive, age-related loss of TRA expression would ultimately be expected to lead to progressive failure of central tolerance, and the release of potentially self-reactive cells into the periphery. The experiments described in this application are designed to test this hypothesis. Using a published computational approach, we will revise our list of TRA expressed by mTEC to include multiple sortable quantitative parameters, and direct (hypertext) links to relevant informatic (body atlas) and genetic (mouse mutant) resources, as well as to predicted MHC class II-binding peptides from these TRA. These will be used to generate 10-15 high-priority candidates for construction of peptide:MHC tetramers that can detect the presence of potentially self-reactive T cells. Such cells will be quantitated in young mice (where tolerance should be highly efficient) and mice of progressively advancing ages. We anticipate that aging will result in the appearance and gradual accumulation of T cells that can recognize, and potentially react against, peripheral self-antigens. These studies thus have the potential to provide a mechanistic explanation for the relationship of aging to autoimmunity. The thymus exhibits rapid atrophy with age, reaching peak size at around puberty, and declining progressively thereafter. Consequently, loss of peripheral self-antigens by the thymus could simply be a secondary response to atrophy. However, the thymus is one of the few adult organs with nascent regenerative potential, and can be completely regrown (albeit transiently) using stimuli such as surgical castration. In the same study mentioned above, we showed that most of the affects of aging on the thymus, including the loss peripheral self-antigen expression, persist in the regrown thymus (which, from the standpoint of its molecular signature, is almost indistinguishable from the aged thymus before regrowth). Thus, the aged, regrown thymus would not only be expected to fail to delete potentially self-reactive cells, but to produce them in even larger numbers, since T cell output from the thymus is proportional to its mass. Since pharmaceutical androgen blockade is being tested for its ability to restore thymic output and immune senescence in the elderly, including in otherwise healthy volunteers, we believe it is important to determine whether the regrown thymus does, in fact, produce an even larger pool of cells that recognize self-antigens, and thus may have the potential to be harmful. This will also be tested in the proposed project.